These days, an increasing amount of information can be obtained in graphical forms, such as weather maps, soil samples, locations of nests in a breeding colony, microscopical slices, satellite images, radar or medical scans and X-ray techniques. "High level" image analysis is concerned with the global interpretation of images, attempting to reduce it to a compact description of the salient features of the scene.This book takes a stochastic approach. It studies Markov object processes, showing that they form a flexible class of models for a range of problems involving the interpretation of spatial data. Applications can be found in statistical physics (under the name of "Gibbs processes"), environmental mapping of diseases, forestry, identification of ore structure in materials science, signal analysis, object recognition, robot vision, and interpretation of images from medical scans or confocal microscopy.
Part 1 Point processes: the Poisson process; finite point processes; interior and exterior conditioning. Part 2 Markov point processes: Ripley-Kelly Markov point processes; marked point processes; nearest-neighbour Markov point processes. Part 3 Statistics for Markov point processes: simulation; parameter estimation. Part 4 Applications: modelling of spatial patterns; higher-level vision.